A significant barrier to effective research , diagnosis, and treatment of stuttering has been the practice of investigators and clinicians to view it as a single disorder -- a pathognomonic monolith -- in spite of its immense diversity. Therefore, the broad objective of the proposed multi-site project is to study the disorder by pursuing delineation of stuttering subtypes. Both a priori subtypes, with particular focus on Persistent and Recovered stuttering, and posteriori subtypes, will be investigated in preschool and early school age children. Our theoretical orientation rests on a dynamic, multifactorial model of stuttering, asserting that stuttering involves complex, multilevel and dynamic epidemiologic, motoric, linguistic, and psychosocial processes that interact to ultimately effect speech production. We posit that stuttering is a developmentally nonlinear disorder expressed as several subtypes that are genetically based and environmentally shaped, reflecting differences in language, motor, and personality/temperament dynamics, manifested in varied symptom patterns. Based on these views, we have generated a set of specific, testable predictions regarding the nature of early childhood stuttering. To sufficiently evaluate our view of stuttering subtypes, complex, multilevel, and dynamic observations and analyses integrating skills in a range of relevant domains (i.e., genetics, speech motor skills, language ability, and personality-temperament variables) is necessary. Toward this end, eleven scholars from six Midwestern universities will collaborate to form four key research branches focusing on: I. Epidemiological and genetic factors, II. Motoric factors, III. Linguistic factors, and IV. Personality/temperament factors. A total of 450 children, 300 who stutter and 150 controls, their parents, and other relatives, will participate. Our experimental strategy is based on a three-part research agenda: Part 1, a core longitudinal study, will map the pathways that preschool children follow over time in stuttering development, evaluating patterns of strength and weakness in performance across a range of areas (motoric, language, personality) as well as monitoring of fundamental epidemiologic and genetic factors. Part 2 will focus on a group of older children with several years of stuttering history to enable immediate study of the later stuttering pathway and compare skill profiles with those of children at stuttering onset. An identical protocol will be followed by all branches and data will be shared and integrated. In Part 3, each center will carry' out branch-specific studies that focus on preliminary data on potentially rich/informative measures concerning risk factors related to the particular domain. By developing profiles of strength and weakness in skills both within and across domains of study, and through advanced statistical modeling techniques, the full set of integrated findings will be utilized to delineate working subtypes of stuttering in young children.